Crop Quality Assesment Through Crop Disease in Agriculture Using Deep Learning Models : A Systematic Review

Abstract

Owing to population growth and changing dietary habits, global food demand has been rising. The demand for staple crops will rise as the population of low-income countries grows. In addition, the change in dietary preferences would be aided by rising income and urbanization. Owing to the rise of crop diseases, agriculture and food systems would be under pressure to fulfil consumer food and nutrient needs, as result crop quality will be decreased and Crop production will be reduced. Crop diseases pose a significant threat to food security, but due to the lack of a critical base in many parts of the world, quickly identifying proofs remains a challenge. In the field of leaf-based image recognition, the development of precise techniques has yielded promising results. The proposed solutions are limited in scale and completely reliant on deep learning models. Convolutional neural networks are proving to be one of the most effective methods for diagnosing and predicting pathogens from crop images. This paper focus on latest neural network methods for processing images, with a focus on crop disease detection and crop quality scaling. First, a look at the data sources, deep learning models/architectures, and various image processing techniques that were used to process the imaging data. Second, the report discussed the findings of a comparison of different current deep learning models, as well as the possible potential of hyperspectral data processing. The aim of this survey's preparation is to enable future research to learn more about deep learning capabilities while detecting plant diseases by improving machine efficiency and accuracy.